use dask threads schedule with default values.

This is faster an reducues the over head of maintaining the dask configuration
chanzuckerberg · Jan 15, 2025 · 1fe16e3 · 1fe16e3
1 parent 093e4ae
commit 1fe16e3
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 9 deletions.
diff --git a/cellxgene_schema_cli/cellxgene_schema/validate.py b/cellxgene_schema_cli/cellxgene_schema/validate.py
@@ -2130,14 +2130,8 @@ def validate(
     validator = Validator(
         ignore_labels=ignore_labels,
     )
-    with dask.config.set(
-        {
-            "num_workers": n_workers,
-            "threads_per_worker": 1,
-            "distributed.worker.memory.limit": "6GB",
-            "scheduler": "threads",
-        }
-    ):
+
+    with dask.config.set({"scheduler": "threads"}):
         validator.validate_adata(h5ad_path)
         logger.info(f"Validation complete in {datetime.now() - start} with status is_valid={validator.is_valid}")
 

diff --git a/cellxgene_schema_cli/requirements.txt b/cellxgene_schema_cli/requirements.txt
@@ -2,7 +2,7 @@ anndata==0.11.2
 cellxgene-ontology-guide==1.3.0 # update before a schema migration
 click<9
 Cython<4
-dask==2024.12.0
+dask[array]==2024.12.0
 numpy<3
 pandas>2,<3
 PyYAML<7